Every run: the agent's full trajectory, its submissions over the 20-hour budget, and the final score.
| # | Model | Reward | Cost | In tok | Out tok | Time | Steps | Submits |
|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 | 0.4716 | $208.91 | 376.5M | 866k | 17h 50m | 1017 | 10 |
| 2 | Claude Fable 5.1 | 0.3532 | $168.41 | 314.0M | 539k | 12h 53m | 608 | 4 |
| 3 | Claude Fable 5.1 | 0.3481 | $106.70 | 193.6M | 429k | 9h 8m | 454 | 13 |
| 4 | Claude Fable 5.1 | 0.3402 | $125.75 | 229.9M | 473k | 10h 25m | 482 | 6 |
| 5 | GPT-5.6 | 0.2171 | $270.86 | 274.0M | 362k | 12h 30m | 806 | 4 |
| 6 | Claude Fable 5.1 | 0.1985 | $113.40 | 205.2M | 463k | 5h 28m | 430 | 3 |
| 7 | GPT-5.6 | 0.1552 | $336.67 | 340.6M | 347k | 11h 28m | 906 | 14 |
| 8 | GPT-5.6 | 0.1222 | $253.61 | 257.1M | 305k | 8h 28m | 760 | 3 |
| 9 | GPT-5.6 | 0.1119 | $309.74 | 312.3M | 342k | 9h 22m | 857 | 12 |
| 10 | GPT-5.6 | 0.1032 | $420.32 | 406.2M | 354k | 11h 35m | 880 | 13 |
| 11 | Grok 4.6 | 0.0526 | $317.62 | 274.4M | 1.8M | 19h 55m | 1078 | 110 |
| 12 | Grok 4.6 | 0.0382 | $286.38 | 260.8M | 1.4M | 19h 57m | 984 | 103 |
| 13 | Grok 4.6 | 0.0273 | $488.84 | 437.7M | 2.0M | 19h 56m | 1536 | 307 |
| 14 | Grok 4.6 | 0.0266 | $454.46 | 403.4M | 2.0M | 19h 58m | 1435 | 75 |
| 15 | Kimi K3 | 0.0249 | $126.07 | 350.4M | 1.1M | 19h 56m | 936 | 7 |
| 16 | Kimi K3 | 0.0214 | $168.60 | 476.4M | 1.3M | 20h 1m | 1079 | 5 |
| 17 | GLM-5.3 | 0.0166 | $91.30 | 316.9M | 852k | 19h 35m | 891 | 7 |
| 18 | Grok 4.6 | 0.0147 | $328.36 | 283.8M | 1.8M | 19h 57m | 1084 | 57 |
| 19 | Gemini 3.7 Flash | 0.0116 | $11.28 | 68.7M | 237k | 4h 7m | 304 | 2 |
| 20 | Kimi K3 | 0.0104 | $167.00 | 460.9M | 1.4M | 20h 1m | 1039 | 1 |
| 21 | Kimi K3 | 0.0079 | $184.18 | 516.4M | 1.3M | 19h 37m | 1094 | 2 |
| 22 | DeepSeek V4 Flash Vision Exp | 0.0069 | $10.96 | 663.9M | 1.1M | 19h 2m | 1527 | 2 |
| 23 | DeepSeek V4 Flash Vision Exp | 0.0065 | $10.60 | 633.4M | 1.1M | 15h 56m | 1516 | 8 |
| 24 | Gemini 3.7 Flash | 0.0064 | $35.75 | 195.3M | 360k | 8h 28m | 555 | 1 |
| 25 | GLM-5.3 | 0.0054 | $111.06 | 362.5M | 755k | 19h 29m | 893 | 10 |
| 26 | Gemini 3.7 Flash | 0.0054 | $6.52 | 51.3M | 224k | 2h 40m | 274 | 2 |
| 27 | GLM-5.3 | 0.0038 | $97.66 | 297.0M | 573k | 19h 4m | 633 | 5 |
| 28 | DeepSeek V4 Flash Vision Exp | 0.0028 | $12.40 | 756.6M | 1.2M | 19h 8m | 1756 | 39 |
| 29 | DeepSeek V4 Flash Vision Exp | 0.0027 | $6.77 | 415.2M | 638k | 18h 28m | 837 | 7 |
| 30 | Gemini 3.7 Flash | 0.0020 | $17.05 | 105.9M | 300k | 5h 32m | 381 | 1 |
| 31 | Qwen3.8-Max | 0.0018 | $183.02 | 558.1M | 1.7M | 19h 56m | 1375 | 1 |
| 32 | Qwen3.8-Max | 0.0016 | $153.43 | 448.8M | 1.7M | 19h 59m | 1202 | 1 |
| 33 | Qwen3.8-Max | 0.0010 | $149.46 | 419.0M | 1.2M | 19h 25m | 993 | 2 |
| 34 | Gemini 3.7 Flash | 0.0009 | $11.28 | 77.4M | 230k | 5h 1m | 369 | 3 |
| 35 | GLM-5.3 | 0.0006 | $94.18 | 300.9M | 621k | 18h 57m | 592 | 10 |
| 36 | Qwen3.8-Max | 0.0001 | $169.64 | 519.3M | 1.6M | 20h 0m | 1239 | 1 |
| 37 | Muse Spark 1.2 | 0.0000 | $7.77 | 40.8M | 287k | 2h 9m | 182 | 5 |
| 38 | Inkling | 0.0000 | $6.72 | 7.0M | 25k | 22m | 138 | 2 |
| 39 | Qwen3.8-Max | 0.0000 | $189.91 | 559.5M | 1.9M | 19h 51m | 1464 | 3 |
| 40 | Kimi K3 | 0.0000 | $182.81 | 498.0M | 1.4M | 20h 1m | 1063 | 2 |
| 41 | Muse Spark 1.2 | 0.0000 | $8.07 | 38.9M | 412k | 3h 7m | 202 | 4 |
| 42 | DeepSeek V4 Flash Vision Exp | 0.0000 | $11.01 | 666.7M | 1.1M | 19h 11m | 1467 | 11 |
| 43 | Muse Spark 1.2 | 0.0000 | $6.30 | 24.2M | 135k | 1h 37m | 159 | 4 |
| 44 | GLM-5.3 | 0.0000 | $103.09 | 375.1M | 554k | 19h 46m | 790 | 12 |
| 45 | Inkling | 0.0000 | $6.59 | 6.8M | 31k | 2h 10m | 146 | 2 |
| 46 | Muse Spark 1.2 | 0.0000† | $2.64 | 12.4M | 42k | 2h 25m | 90 | 2 |
| 47 | Inkling | 0.0000 | $6.57 | 6.8M | 31k | 35m | 156 | 1 |
| 48 | Inkling | 0.0000 | $4.24 | 4.4M | 17k | 41m | 109 | 1 |
| 49 | Inkling | 0.0000 | $8.87 | 9.2M | 43k | 1h 49m | 170 | 2 |
| 50 | Muse Spark 1.2 | 0.0000† | $1.72 | 8.1M | 39k | 1h 17m | 88 | 1 |
† score adjusted by manual review; the trace's Judge tab carries the measured score and the rationale.